SAVOR: Part I

Modern for Data Science

R started in 1993:

`r 2025 - 1993 =` 32 years ago

A lot has happened since !

• base vs tidyverse
• R GUI vs RStudio
• Rmarkdown, and then Quarto

Data Science

Data Science is an emerging field at the crossing of
Statistics, Computer science & Data analysis

source: R for Data Science (2e), Wickham et al.

Course objectives

• be able to successfully import and transform data in (%>% & dplyr)
• be able to choose and implement suitable and beautiful data visualizations (ggplot2)
• be able to have a reproducible workflow through dynamic reporting
• understand the difference and commonalities between:
  • software development
  • data analysis

Course organization

Parts:

1. (brief) recap on basics
2. Dynamic reproducible reporting with Quarto
3. Data manipulation with dplyr
4. Data visualizationwith ggplot2

In each part:

• some key theoretical concepts
• practicals exercise to develop your abilities and your autonomy

General advices

• Google (or any other web search engine) & ChatGPT are your friends !

• DRY vs WET:

  • Don’t Repeat Yourself !
  • or Write Everything Twice (you have time to spare !)

⇒ use function()

brush-up

• RStudio: use up-to-date, modern, tools

• use RStudio projects – always !!!

⇒ live demo

• loop and functions (DRY)

Brush up practical

open SAVOR_practical1.html and follow along…

tidyverse

Tidy + Universe

Hadley Whickam 🫶

tidyverse: a collection of tidy R packages

🌐👉 www.tidyverse.org

tidy data

1. each column represent a different variable
2. each row represent one observation
3. different observation types are stored in different tables (i.e. data.frame)

⇒ tidyverse: a collection of packages for working with/analyzing tidy data

Other ressources

• Posit cheat sheets and webinars
• R for data science (2e), Whickam, Çetinkaya-Rundel & Grolemund 👉🌐
• What they forgot to teach you about R, Brian & Hester 👉🌐
• many more …